Over 230,000 women are estimated to be diagnosed with breast cancer in 2014, and one in five will have Type II diabetes. Poorer outcomes are apparent for patients with both conditions, with a 52% increased risk of overall mortality. Even with such high prevalence of co-occurrence, there is limited information about whether and why outcomes may be worse in women with both breast cancer and diabetes versus only breast cancer; there is even less information for how best to treat these patients considering the interaction of the diseases and their treatments. For example, taxes and carboplatin, adjuvant chemotherapy agents for breast cancer, may worsen diabetic neuropathy. Delayed wound healing in patients with diabetes may impact chemotherapy optimal timing. Corticosteroids to control vomiting during chemotherapy may elevate blood sugar and complicate diabetes management. Women with diabetes use mammography screening less. Metformin for diabetes may improve therapeutic response in breast cancer. Clearly, care decisions are more complex when patients have more than one serious health condition, and their interaction begs for a systematic evaluation. Accordingly, our Aims and associated hypotheses are: 1) to determine how diabetes impacts the breast cancer quality of care; 2) to determine the impact of diabetes-related variations in breast cancer care on recurrence and breast-cancer-specific and overall mortality; and 3) to determine how breast cancer affects diabetes quality of care in breast cancer patients with diabetes. We will collect comprehensive and granular data on a large sample of patients affected by both diseases, and therefore maximize the potential for evaluating the above Aims and Hypotheses. In order to improve care for multimorbid patients and to better understand how and why patients with breast cancer and diabetes have worse outcomes than those without diabetes, we will analyze two complementary cohorts of subjects with breast cancer: the linked SEER-Medicare data including hospital, outpatient, physician, and medication claims for women ages 66-85 years; and richly-detailed clinical data from an integrated delivery system, Group Health in Western Washington State for women ages 50-85 years. Employing innovative statistical methods on two datasets, one with broad generalizability and one with highly-detailed clinical information, we will evaluate what factors most impact health outcomes in breast cancer patients with and without pre-existing diabetes and whether diabetes mediates racial/ethnic inequities in breast cancer treatment and outcomes. The analyses will use innovative statistical methods to tease apart the complexities of simultaneous disease management. This innovative study will help inform optimal care coordination and evidence-based interventions in order to reduce avoidable morbidity and mortality from breast cancer and diabetes. This work also builds a platform for evaluating the interaction between cancer and other burdensome co-existing comorbidities.